During recent years, there has been a great deal of research and development activity directed to making small metallic particles, especially small ferromagnetic metallic particles for use in magnetic tapes and the like. It will be understood that many of the processing improvements developed during work on magnetic particles will also have an advantage in making non-magnetic metal particles. This is generally true to the extent that such parameters as uniform particle size, optimum particle size, a discrete predictable shape, favorable bulk density or surface area, and other such desirable physical properties are advantageous in metallurgical compounding work, in making catalysts, etc. Also, the reproducible, highly ordered morphology of particles produced by the process of the invention is believed to be of general importance. Thus, although the basic inventions described herein were initially reduced to practice during work directed towards making improved magnetic powders, the emphasis on magnetic particles (and the extraordinary properties thereof) in this application should not be interpreted as an intent to waive coverage on other novel powders or on the novel organometallic salt precursors prepared according to the instantly-disclosed process and from which non-magnetic powders can be prepared.
Considerable emphasis on making high performance magnetic powders has focused, over the last ten or fifteen years, on making high-performance cobalt-based metal particles. There are a number of properties such particles should have and they include the following:
High Coercivity and, preferably, the temperature dependence of the coercivity value should be minimized. PA1 High Sigma Value - usually denotes high metal content, i.e. a minimum amount of surface oxide contamination or minimal use of non-magnetic, coercivity-inducing additives into the crystalline structure. PA1 High Squareness -- the squareness of the hysteresis loop should be maximized. Unless otherwise noted, this squareness is measured in a field of 2000 oersteds PA1 Reasonable particle size -- very small acicular particles, say, those having an average length of as low as 0.05 microns will have enormously large surface areas and will be difficult to disperse in desirable polymeric binder media to form coatings required in manufacture of magnetic tape. (On the other hand, for many purposes, it is well to be able to control the process to achieve much smaller particles.) PA1 High Acicularity -- needle-like metal particles have been difficult to produce. Even when highly acicular organometallic salts are used, the resulting metal is usually less acicular than is desired. PA1 Good Particle Size Distribution -- the above-mentioned average particle size will have little value if the particulate product comprises an excessive surface area due to many very small particles. (On the other hand, larger than average particles may cause problems in metallurgical work.) Moreover, a relatively narrow particle size range is believed, on the basis of work described herein, to provide favorable magnetic properties such as an excellent switching-field distribution and avoidance of bothersome D.C. saturation noise effects in magnetic memory processes such as those utilizing magnetic tapes. PA1 Stability over a period of time - earlier work has resulted in making particles that degraded with respect to magnetic and chemical properties in an undesirably short time.
Other parameters will be discussed in the body of this application.
There have been many attempts to produce improved magnetic metallic particles for incorporation into magnetic coatings. Many of them have involved the decomposition of metal carbonyl or other covalently-bonded compounds. For example, see U.S. Pat. Nos. 3,228,882 and 3,228,881 to Harle et al and Thomas, respectively.
Other such processes have involved the use of borohydrides as reducing agents for metal ions in solution and the subsequent precipitation of the metal powder. Such borohydride type processes are disclosed in such patents as U.S. Pat. Nos. 3,206,338 to Miller et al and 3,661,556 to Jolley et al.
More promising work appears to be the formation and reduction of organometallic salts of carboxylic acids such as metal oxalates and the like. Some such work is described in such U.S. Pat. Nos. as 3,186,829 to Landgraf, 3,574,683 to Johnston, and 3,317,574 to Morita et al. The most pertinent of this work appears in the commonly-owned and co-pending application U.S. Ser. No. 228,387 filed on Feb. 22, 1972 by Ehrreich and Reti now U.S. Pat. No. 3,892,673. In this work, the use of a resinous coating envelope over the particles during reduction of an oxalate salt did result in a product having an excellent combination of squareness, coercivity and sigma value.
However, even the powder products produced by the Ehrreich-Reti process have limited commercial utility because of problems associated with stability of sigma value under relatively high humidity and temperature. These powder materials, when incorporated into magnetic tape, also tended to produce a "noise" about certain narrow wavelengths and, as a consequence, was not desirable for some magnetic tape applications.
It will be evident that the discoveries disclosed herein will be applicable generally to the work of those who have produced powders from organometallic salt particles. That process is characterized by a thermal step wherein the organic component of the salt decomposes directly to gaseous by-products leaving a metallic powder residue behind.
Although it had been known to reduce organometallic salts such as oxalates and the like in a thermal reactor for the purpose of obtaining metal particles, the primary concern had been to avoid such high temperatures as will result in excessive sintering of the particle and chemical reactions which will produce excessive oxide content in the finished particle. Thus, for example, control of sintering is one object of the process for coating metal-bearing organometallic particles with an inert material before and during reduction to the metal powder. That process is described in a commonly-owned and co-pending patent application Ser. No. 367,461 filed on June 6, 1973, by Ehrreich and Reti now U.S. Pat. No. 3,843,349. That process is particularly advantageous because it results in particles having surprisingly desirable magnetic properties. In any event, the aforesaid discovery by Ehrreich and Reti that materials coated with resins could result in markedly improved magnetic properties stimulated further work in the development of improved processes with the general intent of further optimizing the potential magnetic properties, especially with respect to high acicularity, freeness from DC noise manifestation when incorporated in magnetic tape.
In a hindsight review of fields of art which have some relationship to the invention, the following art has been uncovered:
U.S. Pat. No. 3,046,158 discloses the use of acicular seed crystals of iron oxide which seeds are subsequently used as growth sites for production of acicular iron oxide. Two aspects of this patent are particularly noted: the growth upon the seed in no way contributes to acicularity. All acicularity is contributed by the seed itself; future growth in no way accentuates or preserves the acicular character independently of the pre-existing acicular nature of Fukada's seed. Secondly, the material being grown on the seed is the magnetic powder, not a non-magnetic precursor thereof comprising only a minor volume of metal.
U.S. Pat. No. 2,558,304 discusses the use of "crystal growth directors" which form complexes with hydrated iron oxide. The growth directors were used during oxidation of ferrous compounds to make iron oxide pigments for use in paint and the like.
In The Physics of Magnetic Recording by C. D. Mee (North Holland Publishing Company, Amsterdam, 1968), it is disclosed that certain acicular magnetic particles were made by Luborsky (at pages 196-198). This epic attempt to provide magnetic particles of ideal properties was thwarted by the necessity of using special techniques which inherently resulted in low sigma values of the products.
U.S. Pat. No. 3,740,266 describes broad range of acicular metal particles; applicants know of no practical way to produce acicular metal particles from most of the alloys described therein. Most of the specific art cited involve use of large quantities of non-ferromagnetic dopes, e.g. 10-40 mole percent of Mn, Zn, Sn, Cu or Pb; or use of alkaline earth metals or use of very small quantities of cobalt.
Other art which, when superficially examined, seems to relate to the product aspect of the invention is the U.S. Pat. No. 3,228,882 to Harle et al. This discloses highly acicular structure but the structure is not formed or acicular metal particles. It is formed of very small, nearly spherical, metal particles which are incorporated in--and separated by--an acicular matrix of resin. Such particles have a very low sigma value (on a volume basis) and cannot be loaded into magnetic tape formulation in suitable quantity.
U.S. Pat. No. 3,661,556 to Jolley et al shows complexing agents in an aqueous medium in which metal powder product is formed by reduction of metal salts such as borohydride salts. Jolley at al disclaim knowledge of how their mechanism works, but suggest it works by preventing the precipitation of ferric oxide during the formation of the metal. That invention teaches nothing about the utility of similar compounds in the formation of an organometallic salt which is to be collected as a solid and only later is used as a precursor to metal formed by a high-temperature reduction process; i.e. that invention teaches nothing about a process of the type to be described below. Moreover, there are inherent problems associated with the Jolley et al process: e.g. lack of high squareness, even when there is substantial accularity, and relatively high non-magnetic-metal content, which prevent the products thereof from having the unusually excellent and novel magnetic properties of the powders of the invention which will be described below.